{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:HKH5JHGKGH2HBY3YQLBTXINGDD","short_pith_number":"pith:HKH5JHGK","schema_version":"1.0","canonical_sha256":"3a8fd49cca31f470e37882c33ba1a618cc88c9d2778cb4126476d1b1b23ce3e6","source":{"kind":"arxiv","id":"2607.25532","version":1},"attestation_state":"computed","paper":{"title":"Entangled by Design: Spurious Intra-Variable Signal Routing in Tabular In-Context Learners","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Athanasios Vlontzos, Bernhard Kainz, Giorgos Papanastasiou, Sotirios Tsaftaris","submitted_at":"2026-07-28T10:17:59Z","abstract_excerpt":"Consider a model trained at a single hospital to predict patient recovery, where the measured feature $X$ bundles the patient's true health signal ($C$) with a systematic artefact from that hospital's equipment ($S$). Within that hospital, the artefact correlates with outcomes through unmeasured confounders such as patient demographics; an in-context learner rationally routes predictions through $S$, not $C$, and fails silently when deployed at a new hospital with different equipment. We formalise this as \\emph{spurious routing in composite representations}: when a feature $X = [C;\\,\\alpha S;\\"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2607.25532","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-28T10:17:59Z","cross_cats_sorted":[],"title_canon_sha256":"bc168e0a28063e9da6ea7bb10c5b4e54e3cb5e5d5ee95640137f20e71aebd9cf","abstract_canon_sha256":"dba3b31827c0e0e3afcc71b91c2a29c2b3d0686d3a7507b6f2c8a3739edf3f57"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-29T01:25:45.375553Z","signature_b64":"UusEMDRd8giZ/1QcG86eVKvuhOizxKLoO/QSyzEAu1U6D8lUjiz1fKTxc8scWIntixL7HtO8OQLsZQhIIo0YAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3a8fd49cca31f470e37882c33ba1a618cc88c9d2778cb4126476d1b1b23ce3e6","last_reissued_at":"2026-07-29T01:25:45.374694Z","signature_status":"signed_v1","first_computed_at":"2026-07-29T01:25:45.374694Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Entangled by Design: Spurious Intra-Variable Signal Routing in Tabular In-Context Learners","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Athanasios Vlontzos, Bernhard Kainz, Giorgos Papanastasiou, Sotirios Tsaftaris","submitted_at":"2026-07-28T10:17:59Z","abstract_excerpt":"Consider a model trained at a single hospital to predict patient recovery, where the measured feature $X$ bundles the patient's true health signal ($C$) with a systematic artefact from that hospital's equipment ($S$). Within that hospital, the artefact correlates with outcomes through unmeasured confounders such as patient demographics; an in-context learner rationally routes predictions through $S$, not $C$, and fails silently when deployed at a new hospital with different equipment. We formalise this as \\emph{spurious routing in composite representations}: when a feature $X = [C;\\,\\alpha S;\\"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.25532","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2607.25532/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2607.25532","created_at":"2026-07-29T01:25:45.375144+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.25532v1","created_at":"2026-07-29T01:25:45.375144+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.25532","created_at":"2026-07-29T01:25:45.375144+00:00"},{"alias_kind":"pith_short_12","alias_value":"HKH5JHGKGH2H","created_at":"2026-07-29T01:25:45.375144+00:00"},{"alias_kind":"pith_short_16","alias_value":"HKH5JHGKGH2HBY3Y","created_at":"2026-07-29T01:25:45.375144+00:00"},{"alias_kind":"pith_short_8","alias_value":"HKH5JHGK","created_at":"2026-07-29T01:25:45.375144+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HKH5JHGKGH2HBY3YQLBTXINGDD","json":"https://pith.science/pith/HKH5JHGKGH2HBY3YQLBTXINGDD.json","graph_json":"https://pith.science/api/pith-number/HKH5JHGKGH2HBY3YQLBTXINGDD/graph.json","events_json":"https://pith.science/api/pith-number/HKH5JHGKGH2HBY3YQLBTXINGDD/events.json","paper":"https://pith.science/paper/HKH5JHGK"},"agent_actions":{"view_html":"https://pith.science/pith/HKH5JHGKGH2HBY3YQLBTXINGDD","download_json":"https://pith.science/pith/HKH5JHGKGH2HBY3YQLBTXINGDD.json","view_paper":"https://pith.science/paper/HKH5JHGK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.25532&json=true","fetch_graph":"https://pith.science/api/pith-number/HKH5JHGKGH2HBY3YQLBTXINGDD/graph.json","fetch_events":"https://pith.science/api/pith-number/HKH5JHGKGH2HBY3YQLBTXINGDD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HKH5JHGKGH2HBY3YQLBTXINGDD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HKH5JHGKGH2HBY3YQLBTXINGDD/action/storage_attestation","attest_author":"https://pith.science/pith/HKH5JHGKGH2HBY3YQLBTXINGDD/action/author_attestation","sign_citation":"https://pith.science/pith/HKH5JHGKGH2HBY3YQLBTXINGDD/action/citation_signature","submit_replication":"https://pith.science/pith/HKH5JHGKGH2HBY3YQLBTXINGDD/action/replication_record"}},"created_at":"2026-07-29T01:25:45.375144+00:00","updated_at":"2026-07-29T01:25:45.375144+00:00"}